A future mother must want the fetus in her womb to be in a healthy condition. A healthy fetus is one that has optimal growth and has sufficient nutrients at the time it is in the womb. The risk of miscarriage and maternal and fetal deaths can be reduced by monitoring fetal health and staying alert when necessary. CTG is carried out in cases where there is a risk of pregnancy and a relatively worrying birth. It will be easier to make further decisions to reduce health risks at birth if the CTG examination shows a good fetal condition. The study aims to study the comparison of different classification algorithms used in determining the exact values of the fetal health datasets used through Knowledge Discovery in Databases (KDD). The researchers used a data collection of 2126 data points from Kaggle’s website that had 22 variables divided into three categories Normal, Suspect, and Pathological. Prolonged decelerations, abnormal short-term variability, and the percentage of time with abnormally long-term variability are the most significant variables. The results of the study were obtained by dividing the data set into training and trial data, which were then divided into three trials. The results showed that the KNN algorithm had the best accuracy value of 91% in the second trial, the SVM algorithm had the most accurate value of 87% in the first and second trials, the Logistic Regression algorithm had the Best Accuracy Value of 84% in the Second Trial, the Naive Bayes algorithm had the Best Accuracy value of 84%, and the Decision Tree Algorithm had 89% in the First and Second Trials.
Copyrights © 2023